You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This specified problem is introduced since #39760. The parser of compiler always tries to parseJSDocNameReference although there might be none. So, for http://lmgtfy.com, it gets stuck at : since it's an UNKNOWN token for current grammar. Then, http becomes a NameReference, while ://lmgtfy.com is comment. However, the whole string is all comment in fact.
@z0gSh1u's analysis is correct. The comment-parsing part of jsdoc parsing skips initial whitespace, which isn't correct for jsdoc where whitespace is usually significant. getJsDocTags then separates the name of a tag from the comment with a space. Unfortunately, it's a fairly large change to fix this.
The output of getJsDocTags is now DisplayParts in 4.3, but it still inserts the space.
Bug Report
🔎 Search Terms
@see
,JsDocTags
🕗 Version & Regression Information
This changed between 3.9 and 4.1.
💻 Code
🙁 Actual behavior
Consider the exported symbol
Foo
above... When callinggetJsDocTags()
on that, the result includes:🙂 Expected behavior
Expected the
@see
tag's text to be represented as-is:The text was updated successfully, but these errors were encountered: